Overview
Charles Manski is Board of Trustees Professor in Economics at Northwestern University and a Research Associate at NBER. He is a leading authority on identification in econometrics, known especially for developing the theory of partial identification — the systematic analysis of what can be learned about a population parameter from data alone, without untestable parametric or distributional assumptions. His work on identification bounds (Manski bounds) provides informative ranges for treatment effects when full point identification is impossible, and his methodological writings on the limits of natural experiments have been influential in calibrating economists' confidence in quasi-experimental estimates.
Key Contributions / Features
- Partial identification and Manski bounds: When treatment assignment is not fully random (or the exclusion restriction is uncertain), sharp bounds on causal parameters can be derived from the observed data distribution alone; these bounds narrow as additional assumptions are imposed, making the dependence of conclusions on assumptions explicit and auditable
- Identification Problems in the Social Sciences (1995): book-length treatment of identification in econometric models; emphasizes that "identification" is a property of the assumed model, not of the data — a model that point-identifies a parameter does so by assumption, not by empirical content
- Critique of natural experiments: Manski has written on the limits of the "as-if randomization" defense of IV instruments; when the exclusion restriction cannot be directly tested, IV estimates are sensitive to plausible departures — a point formalized in his bounds framework and relevant to the debates in Natural Experiments and Instrumental Variables
- Identification of social interactions: work on the "reflection problem" — disentangling individual behavioral responses to peer outcomes from shared contextual factors — foundational to the peer effects literature
Related
Sources